Universal Representation of Image Functions by the Sprecher Construction
Conference paper
Summary
This paper proposes a procedure for representing image functions by a computation in two layers. It is recalled that the general function representation needs more layers than two, using the Stone-Weierstrass theorem for approximation in three layers, and the Kolmogorov theorem for representation in four layers. For achieving representation in two layers only, the requirement on a continuous representation has to removed. The Sprecher construction presented here is a general procedure for yielding such a representation in two layers. It can be used to compress images, to represent pixels and their neighborhoods directly, or to represent image operators.
Preview
Unable to display preview. Download preview PDF.
References
- 1.Neil E. Cotter and Thierry J. Guillerm. The CMAC and a theorem of Kolmogorov. Neural Networks, 5:221–228, 1992.CrossRefGoogle Scholar
- 2.Robert Hecht-Nielsen. Kolmogorov’s mapping neural network existence theorem. In Proceedings of the First International Conference on Neural Networks, volume III, pages 11–13. IEEE Press, New York, 1987.Google Scholar
- 3.K. Hornik, M. Stinchcombe, and H. White. Multilayer feedforward networks are universal approximators. Neural Networks, 2:359–366, 1989.CrossRefGoogle Scholar
- 4.Andrej Nikolajewitsch Kolmogorov. On the representation of continuous functions of several variables by superpositions of continuous functions of one variable and addition. Doklady Akademii Nauk SSSR, 114:679–681, 1957. (in Russian).MathSciNetGoogle Scholar
- 5.Mario Köppen. On the training of a kolmogorov network. In José R. Dorronsoro, editor, Artificial Neural Networks-ICANN 2002, LNCS 2415, pages 474–479, Madrid, Spain, August 2002, 2002. Springer-Verlag Heidelberg.Google Scholar
- 6.V. Kreinovich, H.T. Nguyen, and D.A. Sprecher. Normal forms for fuzzy logic — an application of kolmogorov’s theorem. Technical Report UTEP-CS-96-8, University of Texas at El Paso, Januar 1996.Google Scholar
- 7.H.T. Nguyen and V. Kreinovich. Kolmogorov’s theorem and its impact on soft computing. In Ronald R. Yager and Janusz Kacprzyk, editors, The Ordered Weighted Averaging operators. Theory and Applications., pages 3–17. Kluwer Academic Publishers, 1997.Google Scholar
- 8.D.A. Sprecher. On the structure of continuous functions of several variables. Transcations of the American Mathematical Society, 115(3):340–355, 1965.MATHCrossRefMathSciNetGoogle Scholar
- 9.D.A. Sprecher. A representation theorem for continuous functions of several variables. Proceedings of the American Mathematical Society, 16:200–203, 1965.MATHCrossRefMathSciNetGoogle Scholar
- 10.D.A. Sprecher. A numerical implementation of Kolmogorov’s superpositions I. Neural Networks, 9(5):765–772, 1996.CrossRefGoogle Scholar
- 11.D.A. Sprecher. A numerical implementation of Kolmogorov’s superpositions II. Neural Networks, 10(3):447–457, 1997.CrossRefMathSciNetGoogle Scholar
Copyright information
© Springer-Verlag Berlin Heidelberg 2005